configuration in config.yaml you can edit the file to run with diffrent config, most scripts get the conf file as an argument so it is possible to save current configurations
all scripts will create intermediate .h5ad files to /outputs dir you may need to create one at first run
raw data pipeline until full annotation
- load_sc_data_to_anndata
- pp_adata
- train_scvi_model
- train_scvi_model_on_sub_population
- infer annotation (also do clustering)
annoation of malignant cells are not yet automated, and is done after running the annotation inferance
most analysis is done in notebooks using .h5ad files created by the pipeline